Pérouse de Montclos Marc-Antoine auteur aut IRD text journalArticle eng text/pdf reformatted digital access This article deals with the political issues of excess mortality due 'to conflicts. Based on a case-study from Nigeria, it addresses two main points. The first has to do with the methodological possibilities to measure the intensity and the human impact of armed conflicts. In the context of developing countries where official statistics are not reliable and/or not available because they are not published, it argues that violent deaths are the prime indicators that can be used. The second point addresses the methodological difficulties to define precisely political violence. It underlines the problems of resorting to categories like civilians and combatants, public and private actors, criminal politicians and politicized criminals, etc. In this regard, it appears that crime statistics can also be politicized. Thus it is very necessary to establish a proper methodology before undertaking any assessment of risk and the human impact of armed conflicts. specialized CONFLIT POLITIQUE VIOLENCE MORTALITE TAUX DE MORTALITE TERRORISME MOUVEMENT ISLAMIQUE ISLAM SECTE MESURE DONNEES STATISTIQUES STATISTIQUE CONFLIT ARME SECURITE HUMAINE CONVENTION DE GENEVE NIGERIA 114DROIT2 3 1 85-101 2013 2249-8095 https://www.documentation.ird.fr/hor/fdi:010064364 2249-8095 [F B010064364] https://www.documentation.ird.fr/hor/fdi:010064364 https://horizon.documentation.ird.fr/exl-doc/pleins_textes/divers15-04/010064364.pdf IRD - Base Horizon / Pleins textes 2015-04-24 2026-01-13 fdi:010064364 fre